Nusselt number estimation using a GBR-GSO-based machine learning predictive model in alumina and titania nanofluids in a boiling process

In current study, a flow boiling experimental examination has been performed on water, water-TiO 2 nanofluid, and water-Al 2 O 3 nanofluid and Gradient Boost Regression model was established to predict the Nusselt number of TiO 2 and Al 2 O 3 nanofluids. Nusselt number was calculated by varying heat flux, Reynolds number, and volumetric concentration. Good agreement was shown by water results acquired from experiments with the correlation of Chen and CFD. Results demonstrated that the Nusselt number of TiO 2 and Al 2 O 3 nanofluid was larger than water. Increment in Nusselt number was noticed with the increment in concentration, heat flux, and Reynolds number. It was seen that Al 2 O 3 nanofluid was superior to TiO 2 nanofluid. Highest improvement in the Nusselt number was 37% shown by Al 2 O 3 nanofluid when compared to water. Correlations of Nusselt number were developed for both nanofluids. GBR-GSO machine learning model showed a good agreement with the experimental Nusselt number.